·
13 commits
to main
since this release
OcrNer CLI v1.1.0
Local-first OCR, Named Entity Recognition, and Vision captioning — no cloud APIs required.
Key Features
- Tesseract OCR - Extract text from images with preprocessing (grayscale, contrast, sharpen)
- BERT NER (ONNX) - Named entity recognition running locally via ONNX Runtime
- Florence-2 Vision - Image captioning and vision-based OCR
- ImageSharp Preprocessing - Automatic image enhancement for better OCR accuracy
- Auto-downloads models - Zero manual setup, models download on first use
Downloads
| Platform | Architecture | Download |
|---|---|---|
| Windows | x64 | ocrner-win-x64.zip |
| Windows | ARM64 | ocrner-win-arm64.zip |
| Linux | x64 | ocrner-linux-x64.tar.gz |
| Linux | ARM64 | ocrner-linux-arm64.tar.gz |
| macOS | x64 (Intel) | ocrner-osx-x64.tar.gz |
| macOS | ARM64 (Apple Silicon) | ocrner-osx-arm64.tar.gz |
Quick Start
# Extract and run
./ocrner --help
# OCR an image
./ocrner ocr -f image.png
# Named entity recognition
./ocrner ner -f document.pngFull Changelog: ocrner-v0.0.1...ocrner-v1.1.0